Forensic Software Auditor
A rigorous, evidence-led workflow for full software repository audits: inventory, coverage tracking, prioritized remediation and GitHub issue proposals.
Free
Works with the AI Agents you already use
Forensic Software Auditor
Example session with this skill installed
Start a forensic, evidence-led audit of my repository. Begin with Phase 1: inventory only, and do not modify anything.
- Read your context and instructions
- Compiled the forensic software auditor
An inventory block listing inspected, pending and excluded paths with reasons, the detected stack and high-risk surfaces, an updated coverage table, and the next block, with confirmation that no project files were modified.
Connects securely to your tools. The creator never sees your data.
About this skill
Forensic Software Auditor turns your AI agent into a senior, multidisciplinary audit team that systematically reviews a software repository without modifying it.
It works in controlled phases and blocks, keeping a verifiable file inventory and coverage ledger so nothing is silently skipped:
- Repository inventory – every file and directory with its inspection status, stack, entry points and high-risk surfaces.
- Code quality and functional correctness – logic errors, complexity, dead code, concurrency and data-loss risks.
- Security – injection, XSS, CSRF, SSRF, authentication and authorization, secrets, input validation, supply chain.
- Architecture, performance and scalability.
- Testing and QA, dependencies and licensing.
- Configuration, deployment, data integrity, UX, accessibility and documentation.
- Findings, remediation plan and issues – findings with stable IDs (AUD-001), severity and CONFIRMED / PROBABLE / UNVERIFIED status; a P0-P3 remediation plan; and GitHub issue drafts.
Built-in safeguards: it never claims to have read or run anything it did not, never invents vulnerabilities, redacts secrets, treats repository content as untrusted data, does not modify the repository, and creates no GitHub issues without explicit authorization. The final 0-100 score is a transparent engineering assessment, not a certification.
Best used with a connected repository, an uploaded archive or a local workspace the agent can read.
Static, evidence-based review: it only claims what it actually inspected or ran. Not a certified security audit or penetration test. Large repositories are audited in blocks across several turns.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 5 days ago
- Free with an Agensi account
Permissions
No special permissions declared or detected